Characterization of Two Precursor Polyblends: Polyhydroxyamide and Poly(Amic Acid).
Blends of two precursor polymers. polyhydroxy amide (PHA) and poly(amic acid) (PAA), were studied using differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA). The presence of PHA enhanced thermal and mechanical properties of the polyblends. All of the polyblended films showed large endothermic peaks that decreased monotonically with increasing heat treatment temperature. The cyclization onset temperature ([T.sub.1]), initial decomposition temperature ([T.sub.2]), and weight residue at 900[degrees]C of the polyblends were shown to be in the ranges of 144-146[degrees]C, 532-540[degrees]C, and 44-45%, respectively. Also, the thermal stabilities were enhanced consistently with increasing annealing temperature from 25 to 250[degrees]C. The ultimate strength and initial modulus of the polyblends increased from 84 to 136 MPa and from 2.93 to 5.34 GPa, respectively, with increasing PHA content. Similar to the trend of thermal stability, increasing the annealing temperature of the polyblends inc reased the tensile properties of the films. The observed tensile properties are discussed in terms of the morphology of the fractured films as studied by scanning electron microscopy (SEM). The degree of crystallinity of the polyblends was characterized as a function of heat treatment temperatures by wide angle X-ray diffractometry (WAXD).
Rigid-rod polybenzoxazoles (PBOs) [1-3] and polyimides (PIs) [4-6] are widely used as high-performance polymers and molecular composites due to their excellent thermal stability and mechanical properties.
Earlier we studied poly(amic acid) (PAA) and polyhydroxy amide (PHA) as high temperature precursor polymers [7, 8]. Compared with cyclized polymers, PI and PBO, these precursor polymers have greater solubility and processability. The greater solubility of the polymer precursors makes them a more practical starting point for preparing fibers, films, and molecular composites.
The problem with producing molecular composites is the phase separation of the rigid rod component within polyblends. In order to eliminate the phase separation problem, a polymer precursor is combined with a matrix component and then thermally processed to convert the precursor into a rigid rod cyclized polymer dispersed in the matrix [9-12]. The in situ schemes [10, 12, 13] have the advantage of better compatibility of the precursor polymers with the matrix components over that of fully cyclized PBO or PI.
Kochi et al.  reported one in situ composite using a mixture of PAAs. They showed that high moduli ([greater than] 50 GPa) could be obtained from thin films after cold drawing in the precursor stage. At the same time, Yoon et al.  showed that the chemistry of PAA mixtures is more dynamic than originally assumed. Since amic acid formation is an equilibrium reaction, transamidation can occur when two PAAs or two precursors with very similar structure are combined in solution. Therefore, they can form block or random copolymer structures. Ree et al.  used mixtures of poly(amic acid)s and poly(amic ester)s as in situ composites to eliminate the possibility of amide exchange.
Precursor polymers  have the advantages that they are easier to process, do not require strong solvents, and can adsorb large amounts of heat energy in the cyclization process. Thus if these materials are used in the uncyclized form, they will have increased heat capacity, and when cyclized they will liberate water or a flame retardant, depending upon the design of the chemical groups undergoing the cyclization. Hence, the precursor will be converted to a high temperature heterocyclic polymer possibly containing a flame retardant, and will consume considerable energy in the process [5, 16].
The use of precursor/precursor blends also avoids the processing disadvantages such as high pressures, strong solvents, and equipment erosion that occur when processing the cyclized polymers. However, there have been few studies concerned with how precursor polymer processing variables such as solvent concentration, processing temperature, and heattreatment affect the mechanical properties of precursor/precursor blends.
In studies of the blending process, the effect of heat-treatment on the thermal stability, mechanical properties, and morphology of polyblends has been reported in many papers. Several studies [17-20] have found that heat-treatment improved the thermal and tensile properties. The present work focuses on the thermal and mechanical properties of blends of PAA with PAA and compares the results with that of the pure precursor polymers.
The precursor polymer, PHA, was prepared by solution polymerization  of the 3,3'-dihydroxybenzidine and isophthaloyl dichloride in N,N-dimethylacetamide (DMAc). PAA used for this work was supplied by DuPont as a 15 wt% solution in DMAc. Table 1 shows the general properties of the precursor polymers. The chemical structures of PHA and PAA are as follows:
PHA and PAA in solution were mixed at room temperature. For each blend system, a solution in DMAc was prepared at various concentrations by mixing for one day under vigorous agitation. The concentration of the blended solution was 10% by weight. The polymer solution was coated onto a glass plate and dried in a vacuum oven at 60[degrees]C for a day. Then the blended films, while still on the glass plates, were cleaned in an ultrasonic cleaner 5 times for 30 min each. These films with solvent removed were dried again in a vacuum oven at 60[degrees]C for a day. The film thicknesses were typically in the range of 10-15 microns, and obtained films were light brown for all compositions. For annealing higher temperatures, dark brown films were obtained.
The thermal behavior of the samples were studied by using a DuPont 910 differential scanning calorimeter (DSC) and thermogravimetric analyzer (TGA). DSC and TGA measurements were conducted under a nitrogen atmosphere at a heating rate of 20[degrees]C/min.
Wide angle X-ray diffractograms were obtained on a JEOL JDX-8D instrument using Ni-filtered Cu-K[alpha] radiation. The scan speed was 4[degrees]/min.
All samples of the blended films were tested in tensile mode on an Instron mechanical tester, model No. 5564. The specimens were prepared by cutting 5 by 70 mm long strips. Tensile specimens were tested using a crosshead speed of 5 mm/min. A minimum of six samples was tested for each composition and average value and standard deviations were calculated from the data. Morphology of the fracture surfaces of blended films were prepared by fracture in liquid nitrogen followed by coating with gold on a sputter coater. Fracture surfaces were viewed on a Hitachi-S 2400 scanning electron microscope (SEM).
RESULTS AND DISCUSSION
The thermal properties of the blended films were investigated for films annealed for 30 min in a temperature range of 25-250[degrees]C. The results are summarized in Table 2. Blends of PHA/PAA at room temperature showed two transitions in the DSC, one showed at 179[degrees]C and the other showed at 316--318[degrees]C for PHA contents in the range of 25-75%. These values are not different than the pure precursor polymers. This indicates that the two precursors are not miscible with each other at room temperature. (see Fig. 1).
For samples annealed for 30 min in the range of 100-250[degrees]C, there was only one peak in the DSC that shifted to higher temperatures with increasing annealing temperature. In particular, the higher temperature transition of the 100/0 (pure PHA) occurred from 319 to 334[degrees]C with increasing heat treatment up to 100[degrees]C and then was constant above 100[degrees]C. But for the case of 0/100 (pure PAA) composition, the endothermic minimum point increased from 178 to 235[degrees]C with increasing annealing temperature from 25 to 150[degrees]C, respectively, and then the endothermic peak disappeared for samples annealed above 250[degrees]C as seen in Fig. 3. This suggests that PHA was not affected significantly depending upon annealing, whereas PAA becomes fully cyclized to the polyimide at 250[degrees]C. The corresponding DSC profiles are shown in Figs. 1, 2, and 3 for comparison.
The endothermic enthalpys ([delta]H) of the two precursors are much larger than those for other thermoplastic polymers. Unlike other polymers, large heat capacities were shown to be 221.1 J/g and 122.1 J/g for the PHA and PAA, respectively. One advantage of these large endothermic enthalpies ([delta]H) is that the reaction acts as a heat sink to slow down combustion (15, 16). With increasing annealing temperature from 25 to 250[degrees]C, the endotherms ([delta]H) for the pure precursors and blended composites become smaller than those of the unannealed samples, as listed in Table 2. This can be attributed to the hetero-cyclization reaction with increasing heat treatment that usually occurs in precursor polymers and to the release of small molecule by-products (15).
Thermogravimetric analysis of precursor polyblends revealed significant weight losses due to cyclization ([T.sub.1]) at 255-400[degrees]C (11%) for pure PHA and 145-250[degrees]C (9%) for pure PAA as shown in Fig. 4. The weight loss ([T.sub.1]) of the precursor blends were 18-20% and occurred between 140 and 300[degrees]C (Table 3). The first-stage weight reduction ([T.sub.1]) is brought about by the releasing of water molecules during heterocyclization. These observed weight losses apparently correspond to endothermic transitions in the DSC thermograms as shown in Fig. 1. Initial decomposition onset points ([T.sub.2]) of the polyblends increased from 532 to 540[degrees]C with increasing the PHA content (Table 3).
Thermogravimetric analysis results of precursor polyblends (50/50 PHA/PAA) annealed at different temperatures are listed in Table 4. The TGA data indicates that the thermal stabilities increased steadily with increasing annealing temperature. However, it is noteworthy that [T.sub.1] increased from 144 to 272[degrees]C with increasing annealing temperature from 25 to 250[degrees]C. Also the weight residue at 900[degrees]C increased from 45 to 57% as annealing temperature increased (Fig. 5). Here the good thermal stability of precursor blends, as previously shown in Table 3 and Fig. 4, is also evident.
Tensile properties of the pure precursors and their blends are shown in Table 5. The tensile strengths of the pure PHA and PAA at room temperature were found to be 137 and 84 MPa, respectively. There was no improvement in tensile strength on adding 25% PHA to a blend. But the strengths of the 50/50 and 75/25 PHA/PAA were 89 and 136 MPa, respectively.
At low weight percentages of PHA in the polyblends, the blends exhibited tensile strengths close to that of the pure PAA, as PAA was the dominant phase or matrix. When the PHA loading was increased to 75%, there was an obvious increase in the tensile strength, because PHA is now the dominant phase and PHA has a higher tensile strength than PAA (see Fig. 6).
Ultimate strength of polyblends annealed at 150[degree]C showed similar tensile behavior compared with samples at room temperature. The higher % of PHA leads to the higher tensile strengths of the blends. In particular, the strength of the 75/25 PHA/PAA is higher than that of the pure PHA (145 MPa). The reason is that the partial cyclization of the PAA phase in a blend may affect the strength of the composite.
For the samples annealed at 250[degrees], the strength of the pure PAA and their blends is significantly improved compared with pure PHA. The strengths of pure PAA and pure PHA were 185 MPa and 143 MPa, respectively, and 169 MPa for the three polyblends. As described above, the PAA in the blend has fully cyclized to polyimide at 250[degrees]C, whereas PHA has not been affected significantly at the same annealing temperature.
Similar to the trend in strength, the initial modulus of the blends stayed constant up to 50% PHA content for samples annealed at room temperature (2.93-2.96 GPa). While, above 50 wt% in PHA the blend showed a tremendous improvement in the initial modulus as shown in Table 5 and Fig. 7. This is probably due to the rigid rod aromatic character of the PHA in the polyblends. It is evident that the higher content of rigid rod PHA plays an important role in this PHA/PAA system.
At constant annealing times (3 hrs), as shown in Table 5, the initial modulus of pure PHA and pure PAA increased linearly from 5.94 to 6.27 GPa and from 2.96 to 3.01 GPa, respectively, with increasing annealing temperature from 25 to 250[degrees]C. The effect of heat treatment on the initial modulus is shown in Fig. 7. For the samples annealed at 250[degrees]C, initial modulus of the polyblends is nearly the same as that of PAA up to an initial 50 wt% PHA, further addition shows a dramatic increase in modulus to 100 wt% PHA.
Addition of PHA showed a decrease from 72% to 9% in elongation at break for room temperature annealed samples as presented in Table 5. When annealing temperature increased from 25 to 250[degrees]C, elongation at break of pure PAA decreased from 72% to 61%. On the other hand, there was no change in the elongation at break for polyblends with various PHA contents, when annealed at 150[degrees]C and 250[degrees]C. It was found to be near 4%. Upon heat treatment, the blends showed higher strength and moduli.
Overall, the tensile properties of precursor polyblends were significantly improved by the addition of PHA.
The morphology of fractured specimens was examined to investigate the relationship between the mechanical properties and the resulting microstructure of PAA based composites. SEM micrographs of fracture surfaces of pure precursors and their precursor blends are shown in Figs. 8 and 9. Figure 8a and e show the morphologies of the pure PHA and PAA, respectively. The PHA phase in all blends is typically 50 to 90 nm in diameter and appears to be uniformly distributed over the fracture surface (Fig. 8b-d). The 25/75 PHA/PAA blend shows fine PHA particles 50-70 rim in diameter (Fig. 8d) and the 50/50 PHA/PAA blend also shows a fine dispersion with a domain size of 70-90 rim in diameter. More agglomerated PHA particles due to increasing PHA content were shown in Fig. 8c. When the PHA content increased to 75 wt% however, the spherical domain turned to fibril domains of PHA, which was similar to pure PHA with an average diameter of 40-50 rim as shown in Fig. 8b. This is in agreement with the mechanical property trend s, in which the tensile strength and modulus drastically improved when the PHA content was 75 wt% (see Table 5). In Fig. 8, the presence of any open space around the PHA domains was not observed in the SEM micrographs. However, occurrence of no open circles for the blends does not imply good interfacial adhesion between PHA and PAA. At sufficiently high temperatures under appropriate conditions these polyblends show transamidation reactions below the crystalline melting point. In the course of these reaction, polyblend can produce a compatible blend [6, 11]. Compatibility and transamidation of our system at various annealing temperatures will be reported later .
Also, SEM micrographs of the 50/50 PHA/PAA polyblends with heat treatment at different annealing temperatures are shown in Fig. 9a-f. For the blend annealed at 100[degrees]C for 30 mm (Fig. 9b), no changes were observed when compared with the unannealed sample. When annealed at 250[degrees]C, finely dispersed fibril of the PHA phase were present having diameters of 40-50 nm (Fig. 9c and d). Increasing the annealing temperature to 350[degrees]C, the 50/50 blend showed a dimple-like structure probably due to partially cyclization of the PHA (Fig. 9e). Figure 9f shows a 50/50 PHA/PAA sample annealed at 450[degrees], which is the cyclization temperature of PHA to form the fully heterocyclic polymer, PBO.
Degree of Crystallinity
A rough estimate of the degree of crystallinity of the films was made from the areas of the crystalline and amorphous diffractions (22). Typical X-ray diffractograms of the composites at various annleaing temperatures are shown in Fig. 10. Diffraction from the 25/75 PHA/PAA is similar to that of the pure PAA. However, when the PHA content was increased to 75%, the diffraction peaks due to pure PAA at 2[theta] equal to 10[degrees] and 15[degrees] disappeared, while, a medium new diffraction was observed at 2[theta] = 17[degrees] in pure PHA. A sharp and strong diffraction at 2[theta] equal to 8[degrees] did not change regardless of the PHA content in all the polyblends.
Figure 11 shows the X-ray diffractograms of 50/50 PHA/PAA at different annealing temperatures. Diffractions from blends when annealed from 100 to 250[degrees]C are similar to that of the unannealed blend. It is assumed that the heterocyclization of PAA precursor did not contribute to the crystallinity changes of the composites. For annealing temperatures up to 450[degrees]C, two new diffractions were observed at 2[theta] equal to 13[degrees] and 17[degrees] due to the complete cyclization of PHA.
Degree of crystallinity of the films increased from 10 to 30% with increasing PHA content for unannealed samples. For the 50/50 PHA/PAA. the crystallinity increased linearly from 28 to 50% with increasing annealing temperature from 100 to 450[degrees]C (Table 6).
From the relationship between degree of crystallinity and annealing temperature, it can be seen that the degree of crystallinity increased with the annealing temperature due to the heterocyclization of the precursors. This result shows the correlation between the morphology of the blends and measured tensile properties (Table 5).
This study attempts to illuminate the thermal and mechanical properties of precursor composites from a PBO precursor, PHA, and a PI precursor, PAA, through solution blending. Moreover, the different thermal histories used to obtain the results produces different morphologies and consequently different mechanical behavior in the composite materials.
Precursor polyblends showed large thermal endothermic peaks. In addition, they showed cyclization acts as a heat sink as well as releasing flame quenching molecules such as water upon heating. But those heterocyclic endotherms decreased with increasing annealing temperature. Annealing enhanced the thermal properties, degree of crystallinity, and the mechanical properties of the precursor polyblends. From SEM micrographs of the blended films, a highly dispersed PHA phase was observed in a matrix PAA phase with domain size of 40-90 nm in diameter.
This work was supported by the Kumoh National University of Technology Research Grant (1999). And we thank the U.S. Federal Aviation Administration (FAA).
(*.) To whom correspondence should be addressed.
(1.) J. F. Wo1f, B. H. Loo, and F. E. Arnold, Macromolecules, 14, 915 (1981).
(2.) T. D. Aklnseye, I. I, Harruna, and K. B. Bota, Polymer, 38, 2507 (1997).
(3.) Y. Imai, K Uno, and Y. Iwakura, Makromol. Chem., 83, 179 (1965).
(4.) C. E. Sroog, Ptog. Polym. Sct, 16, 561 (1991).
(5.) K. L. Mittal, Polyimides: Synthesis, Characterization, and Applications, Plenum Press, New York (1984).
(6.) C. Feger, M. M. Khojasteh, and M. S. Htoo, Advances in Polyimide Science and Technology, Technimic, Lancaster, PA (1993).
(7.) J.-H. Chang, M. J. Chen, and R. J. Farris, Polymer., submitted for publication.
(8.) J.-H. Chang and R J. Farris, Polym. Eng. Sci., in press.
(9.) F. W. Harris, S. L.-C. Hsu, and C. C. Tso, Polym. Prepr., 31(1). 342 (1990).
(10.) R. Yokota, R. Horiuchi, M. Koichi, H. Soma, and I. Mita, J. Polym. Sci, Polym. Lett Ed., 26, 215 (1988).
(11.) M. Ree, D. Y. Yoon, and W. Volksen, Polym. Prepr., 31(1), 613 (1990).
(12.) M. Ree and D. Y. Yoon, ACS Polym. Mater. Sci. Eng., 65, 48 (1991).
(13.) M. Ree, T. J. Shin, S. I. Kim, S. H. Woo, and D. Y. Yoon, Polymer, 39, 2521 (1998).
(14.) D. Y. Yoon, M. Ree, W,. Volksen, D. Hofer, L. Depero, and W. Parrish. 3rd International Conference on Polyimides, Mid-Hudson Section of SPE, p. 1 (1988).
(15.) H. J. O'donnell and D. G. Baird, Polym. Eng. Sci, 36, 963 (1996).
(16.) J.-H. Chang and B.-W. Jo, J. Appl. Polym. Sci, 60. 939 (1996).
(17.) T. M. Malik, P. J. Carreau, and N. Chapleau, Polym. Eng. Sci., 29, 600 (1989).
(18.) D. Dutta, H. Fruitwala, A. Kohli, and R. A. Weiss, Polym. Eng. Sci., 30, 1005(1990).
(19.) T. D. Flaim, C. P. Ho, and M. J. Pfeiffer, The Fourth International Conference on Polyimides, p. III-17, Society of Plastic Engineers (1991).
(20.) T. Kubota and R. Nakanishi, J. Polym. Sci. B, 2, 655 (1964).
(21.) J.-H. Chang and R. J. Farris, unpublished results.
(22.) J. F. Rabek, Experimental Methods in Polymer Chemistry, p. 507, John Wiley and Sons, New York (1980).
General Properties of PHA and PAA. Polymer IV [a] T [b], [degrees]C [delta]H [c], J/g [T.sub.d] [d], [degrees]C PHA 1.32 319 221.1 577 PHA 2.01 178 122.1 533 (a.)Inherent viscosity of the PHA was measured at 30[degrees]C at 0.2 g/dL solution in NMP. Intrinsic viscosity of the PAA was measured at 30[degrees]C by using 15 wt% solution in N,N-dimethyl acetamide (DMAc). (b.)Minimum point in endothermic curve. (c.)Endothermic enthalpy. (d.)Initial decomposition temperature. Thermal Properties of the Polyblend Films Annealed for 30 Minutes at Different Heat Treatment Temperatures. Annl. Temp [degrees]C PHA/PAA, wt% [T.sub.c] [a], [degrees]C 25 (r.t.) 100/0 319 75/25 179 316 12.4 50/50 179 318 54.7 25/75 179 318 107.0 0/100 178 122.1 100 100/0 334 75/25 262 50/50 254 25/75 238 0/100 198 150 100/0 335 75/25 308 50/50 268 25/75 259 0/100 235 250 100/0 334 75/25 302 50/50 299 25/75 303 0/100 -- Annl. Temp [degrees]C [delta]H [b], J/g 25 (r.t.) 221.1 100.8 66.0 10.7 100 220.4 215.8 200.7 93.5 79.9 150 218.8 210.8 191.3 80.4 42.9 250 184.7 144.9 73.6 36.3 0 (a.)Minimum point in endothermic curve. (b.)Endothermic enthalpy. Thermogravimetric Analysis of the Polyblend Films. PHA/PAA [T.sub.1] [a] Wt. Red. [b] [T.sub.2] [c] [wt.sup.R] 900 [d] wt% [degrees]C % [degrees]C % 100/0 255 11 (10) [e] 577 45 75/25 145 20 540 44 50/50 144 18 537 45 25/75 146 18 532 45 0/100 145 9 (9) 533 41 (a.)1st weight reduction onset temperature. (b.)Weight reduction between 140[degrees]C and 400[degrees]C in TGA thermogram. (c.)2nd weight reduction onset temperature. (d.)Weight percent of residue at 900[degrees]C. (e.)Values in parentheses represent calculated value. Thermogravimetric Analysis of the 50/50 PHA/PAA Films Annealed for 30 Minutes at Different Temperatures. Annl. Temp. [T.sub.1] [a] Wt. Red. [b] [T.sub.2] [c] [wt.sup.R] 900 [d] [degrees]C [degrees]C % [degrees]C % 25(r.t.) 144 18 537 45 100 172 15 540 49 150 223 6 542 54 250 272 3 531 57 (a.)1st weight reduction onset temperature. (b.)Weight reduction between 140[degrees]C and 400[degrees]C in TGA thermogram. (c.)2nd weight reduction onset temperature. (d.)Weight percent of residue at 900[degrees]C. Tensile Properties of Polyblended Films Annealed for 3 Hours at Several Different Temperatures. Annl. Temp. PHA/PAA Ult. Str. Ini. Modu. E.B. [a] [degrees]C wt% MPa GPa % 25 (r.t.) 100/0 137 5.94 9 75/25 136 5.34 27 50/50 89 2.94 42 25/75 84 2.93 58 0/100 84 2.96 72 150 100/0 137 5.98 4 75/25 145 3.81 5 50/50 100 3.08 4 25/75 94 3.04 5 0/100 95 2.98 63 250 100/0 143 6.27 4 75/25 169 3.18 4 50/50 169 2.98 4 25/75 169 3.05 4 0/10 185 3.01 61 (a.)Percent elongation at break. Degree of Crystallinity of Pure Precursors and Their Polyblends at Different Annealing Temperatures. Annl. Temp., [degrees]C PHA/PAA, wt% D.C. [a] % 25 (r.t.) 100/0 30 75/25 23 50/50 19 25/75 12 0/100 10 100 50/50 28 150 33 250 37 350 43 450 50 (a.)Degree of crystallinity.
|Printer friendly Cite/link Email Feedback|
|Author:||CHANG, JIN-HAE; FARRIS, RICHARD J.|
|Publication:||Polymer Engineering and Science|
|Article Type:||Brief Article|
|Date:||Feb 1, 2000|
|Previous Article:||Effects of Gauge Length and Strain Rate on Fracture Toughness of Polyethylene Terephthalate Glycol (PETG) Film Using The Essential Work of Fracture...|
|Next Article:||On the Effect of Crystallinity on the Elastic Properties of Semicrystalline Polyethylene.|